CN104503390B - Method for achieving automatic recognition of main station through industrial Ethernet integrated slave station - Google Patents

Abstract

The invention discloses a method for an integrated industrial Ethernet slave station to independently identify a master station, which is applied to an integrated slave station embedded with a Linux operating system. This method is based on the premise that the master station will send identification information to the slave station before the communication between the industrial Ethernet master and slave stations. Identify the information, and distinguish and judge the captured identification information, so as to determine the type of the master station and load the corresponding slave station program. Through this method, the industrial Ethernet integrated slave station can automatically identify the type of the master station connected to it, and complete the automatic loading and running of the corresponding slave station program.

Description

A method for autonomous identification of master station by integrated industrial Ethernet slave station

technical field

The invention belongs to the field of industrial control and relates to an industrial control site. An integrated industrial Ethernet slave station running an embedded Linux operating system can independently identify the type of the master station connected to it when communicating with different types of industrial Ethernet master stations. Select and run the corresponding slave station program.

Background technique

The promotion of Ethernet technology to industrial sites has greatly promoted the improvement of production efficiency, product optimization and many other aspects. At present, there are many kinds of industrial Ethernet standards in the world, each of which has its own advantages. In a certain production process, it is often necessary to use several industrial Ethernet standards together to generate greater benefits. Therefore, it is more popular to use the integrated industrial Ethernet slave station in the industrial control field. However, the selection and operation of the slave station program in the integrated slave station has been using the traditional manual loading operation, that is, to determine the nature of the integrated slave station before powering on. Bring a lot of inconvenience to large-scale industrial production. At present, there is an urgent need for a method that can automatically identify the type of the master station connected to the integrated slave station, and automatically run the corresponding slave station program.

Contents of the invention

The purpose of the present invention is to solve the above-mentioned problems existing in the prior art, and provide a method in which an integrated industrial Ethernet slave station can identify the type of the master station and independently select, load and run the corresponding slave station program.

The method for autonomously identifying the master station of the integrated industrial Ethernet slave station provided by the present invention is applied to the industrial control site, and the integrated industrial Ethernet slave station running an embedded Linux operating system is hereinafter referred to as the integrated slave station; the integrated slave station Integrates Modbus-TCP and POWERLINK slave station programs; this method determines the type of master station connected to the integrated slave station by analyzing and judging the master station identification information from the communication network, so as to select the corresponding integrated slave station program for loading and running, It realizes the automatic identification of the master station type by the integrated slave station and the automatic loading of the integrated slave station program, getting rid of the traditional manual loading method;

The specific implementation steps of this method are as follows:

Step 1, capture the identification data packet sent by the master station in the industrial Ethernet communication network;

Step 1.1, start the Linux operating system in the integrated slave station;

Step 1.2: Set the working mode of the Ethernet chip of the integrated slave station to the promiscuous mode, so that the integrated slave station will not only accept the data packets with its destination address, but all the data packets passing through it, which is convenient for the following to get the master station to send The incoming identification information packet.

Step 1.3, create a raw socket file descriptor through the socket socket function;

Step 1.4, use the recv() function to receive data packets from the communication network, and assign the received data packets to a character pointer variable for subsequent analysis.

Step 2, analyzing the data packets in the captured communication network;

Step 2.1: Divide the format of the data packet captured in the first step, because there is no division format when the captured data packet is assigned to a variable, and the data packet is the Ethernet header from left to right , IP packet header, TCP packet header or UDP packet header, you only need to know the Ethernet packet header when distinguishing the data packets sent by the Modbus-TCP master station and the data packets sent by the POWERLINK master station. Forcibly transform the captured data packet into a struct ether_header structure, then the Ethernet header of the data packet will be preserved.

Step 2.2. Determine the ether_type member variable in the struct ether_header structure. If the value of the variable is 0x0806, it can be determined that the data packet comes from the Modbus-TCP master station, and the Modbus-TCP master station is connected to the integrated slave station. End step 2 , go to step 3; if the value of ether_type is not 0x0806, continue to analyze the data packet.

Step 2.3. Continue to judge the ether_type member variable in the struct ether_header structure. If the value of the variable is 0x88ab, it can be determined that the data packet comes from the POWERLINK master station, and the POWERLINK master station is connected to the integrated slave station. End step 2 and enter step 2. Step 3; If the value of ether_type is not 0x88ab, continue to analyze the data packet.

Step 2.4: Discard the captured data packets, return to step 1, and recapture other data packets in the communication network.

Step 3, according to the analysis result of step 2, load the corresponding slave station program;

In this step, depending on the analysis and judgment results of the ether_type member variable in step 2, choose to load and run different industrial Ethernet slave programs, or not to load and run:

Step 3.1, for the Modbus-TCP master station, Modbus-TCP is an executable program under Linux, and the operation of the Modbus-TCP slave station program is completed through the system() function in the executable file;

Step 3.2. For the POWERLINK master station, the POWERLINK slave station program code is cross-compiled to generate a Linux kernel module, which needs to be loaded into the Linux kernel through the insmod command to run; here, by writing a shell script file, in the executable program Execute the script file by executing system() to complete the loading and running of the POWERLINK slave program.

Advantages and beneficial effects of the present invention

It can be seen from the technical solution adopted in the present invention that this method can realize that the integrated industrial Ethernet slave station can judge and identify the master station connected to it, and load the corresponding slave station program, establish a connection with the master station, and complete the follow-up communication. Task. It gets rid of the problems of heavy workload and many uncertain factors brought about by the traditional manual selection method. The present method is convenient, safe and reliable, and has higher practical value.

Description of drawings

Figure 1 is a flow chart of packet capture;

Figure 2 is an analysis diagram of the Ethernet control chip settings;

Figure 3 is the message analysis diagram of Modbus-TCP and POWERLINK master station;

Figure 4 is an analysis diagram of the Ethernet header of Modbus-TCP and POWERLINK packets;

Fig. 5 is a flow chart of packet analysis.

detailed description

The integrated industrial Ethernet slave station provided by the present invention independently identifies the method for the master station, and the specific steps are as follows:

Step 1, capture the identification data packet sent by the master station in the industrial Ethernet communication network;

Figure 1 shows how to run a program to capture data packets on the communication network under the embedded Linux operating system of the integrated slave station.

First start the Linux operating system, provide the working environment of the software program, and load the af_packet.ko original socket function module. The general embedded Linux operating system often does not have this module, and it needs to be loaded into the kernel by itself. Next, set the working mode of the Ethernet chip connected to the integrated slave station to promiscuous mode, because the default working mode of the Ethernet chip is direct mode or broadcast mode. In the default working mode, the Ethernet control chip will judge each Whether the destination of the physical data frame is the address of this station, if not, it will be discarded. However, the integrated slave station needs to complete the capture and analysis of the data packet with identification information sent by the master station in the communication network before the slave station program runs. Therefore, it is necessary to set the adapter network card to promiscuous mode, so that the purpose of monitoring and capturing network information can be achieved. The modification of the Ethernet card mode is completed through the command "ifconfig eth0promisc". The upper and lower parts of Figure 2 show the status of the Ethernet control chip before and after setting.

Step 2, analyzing the data packets in the captured communication network;

Figure 3 and Figure 4 involved here are all data packets from the communication network captured by the wireshark packet capture tool.

The contents marked in the boxes in Figure 3 are the identification information packets sent by the Modbus-TCP and POWERLINK industrial Ethernet master stations. Observation and comparison show that there are differences in the protocol type (protocol option) between the two. Modbus-TCP is ARP, and POWERLINK It is epl, which is used as the district mark.

Figure 4 is an analysis diagram of the Ethernet packet headers of these two protocols. The method of the invention analyzes the captured data packets, and the involved content is in the data link layer, and mainly judges the Ethernet message header, and does not analyze and judge the upper network including the network layer and the transmission layer. Modbus-TCP and POWERLINK are two industrial Ethernet communication protocols, both of which use the Ethernet II frame format. The reason why it is highlighted here is the Ethernet II type data frame used, because it is a type option after 12 bytes, and The two protocols Modbus-TCP and POWERLINK are distinguished here. The Modbus-TCP protocol is an industrial Ethernet based on TCP/IP, and the content of this item in its Ethernet message header is 0x0806; the POWERLINK protocol is not an Ethernet based on TCP/IP, and it has its own data link layer protocol standard. Therefore, the type content of its Ethernet packet header is 0x88ab. Therefore, the type of the communication master station connected to the integrated slave node can be determined by judging the Ethernet data packet type of the captured data packet.

FIG. 5 is a flow chart of analyzing and judging captured data packets. First of all, according to the form of the data packet, divide the format of the captured data packet, because the captured data packet is not divided into a format for a certain variable, and the data packet from left to right is the Ethernet packet header, IP packet header, TCP packet header or UDP packet header, according to the above analysis, it can be seen that only need to know the Ethernet packet header to distinguish Modbus-TCP and POWERLINK industrial Ethernet. Assign the data packet to a character pointer variable, and force this variable into a struct ether_header structure, then the Ethernet header of the data packet will be preserved. The struct ether_header structure is the function body form of the Ethernet packet header, and its specific composition is as follows:

Then, judge the ether_type item in the Ethernet packet header through the judgment statement. According to the message analysis in Figure 3, if the type of the Ethernet header is 0x0806, it is the identification information of the Modbus-TCP master station, then load the Modbus-TCP slave station program; if the type of Ethernet header is 0x88ab, which is the identification information of the POWERLINK master station, then load the POWERLINK slave station program; if it is not one of these two, then it is a garbage packet, discard it, and Return to step 1 through the while function, and recapture the analysis data package.

Step 3, according to the analysis results of step 2, load the corresponding slave station program

In this step, depending on the analysis and judgment results of the ether_type member variable in step 2, choose to load and run different industrial Ethernet slave programs, or not to load and run. Due to the different characteristics of the slave station programs of Modbus-TCP and POWERLINK, their operation methods are also different. As an executable program under Linux, Modbus-TCP uses the system() function in the executable file to complete the operation of the Modbus-TCP slave station program; and the POWERLINK slave station program code is cross-compiled to generate a Linux kernel module, which needs to pass The insmod command is loaded into the Linux kernel to run, but the insmod command cannot be executed from the executable file to load the kernel module. Here, by writing a shell script file, the script file is executed by executing system() in the executable program , thus indirectly completing the loading and running of the POWERLINK slave program.

Claims (1)

1. a kind of method of integrated EPA slave station autonomous classification main website, the method is applied at Industry Control scene, operation There are the integrated EPA slave station of built-in Linux operating system, hereinafter referred to as integrated slave station；Described integrated slave station is integrated Modbus-TCP and POWERLINK slave station programs；The method judges to recognize letter from the main website in communication network by analysis Cease to determine the main website type being connected with integrated slave station, so as to select the corresponding integrated slave station program of load operating, realize Integrated slave station has broken away from the side of traditional manual loading to the autonomous classification of main website type and the automatic loading of integrated slave station program Method；The method to implement step as follows：

The identification data bag that main website sends in 1st step, crawl EPA communication network；

1.1st step, the (SuSE) Linux OS started in integrated slave station；

1.2nd step, the mode of operation of the integrated slave station Ethernet chip of setting are promiscuous mode, and so integrated slave station would not only connect It is its packet by destination address, but all by its packet, it is convenient to obtain the identification information number that main website sends According to bag；

1.3rd step, by socket socket functions create a filec descriptor for raw socket；

1.4th step, received from the packet in communication network using recv () function, and the packet assignment that will be received It is convenient that subsequently it is analyzed in a character type pointer variable；

Packet in the communication network that 2nd step, analysis are grabbed；

2.1st step, the packet that will be grabbed in the 1st step divide form, because the packet for being captured is assigned to a certain variable When, be not divide form, packet be followed successively by from left to right Ethernet heading, IP headings, TCP message head or UDP headings, only need to know when packet is sent out by differentiation Modbus-TCP main websites and packet is sent out by POWERLINK main websites Ethernet heading.The variable that packet is replicated is forced to be converted into structether_header structures, then The Ethernet heading of packet is just retained；

2.2nd step, the ether_type member variables judged in struct ether_header structures, if the value of variable Packet is then can determine that for 0x0806 from Modbus-TCP main websites, what is be connected with integrated slave station is Modbus-TCP main websites, knot The step of beam the 2nd, into the 3rd step；It is not 0x0806 if the value of ether_type, continues analyze data bag；

2.3rd step, continuation judge the ether_type member variables in struct ether_header structures, if variable Value then can determine packet from POWERLINK main websites for 0x88ab, that be connected with integrated slave station is POWERLINK master Stand, terminate the 2nd step, into the 3rd step；It is not 0x88ab if the value of ether_type, continues analyze data bag；

2.4th step, captured packet is abandoned, return to the 1st step, other packets in communication network are captured again；

3rd step, the analysis result according to the 2nd step, load corresponding slave station program；

This step is different according to the analysis judged result in the 2nd step to ether_type member variables, and selection load operating is different EPA slave station program, or be not loaded with operation：

3.1st step, for Modbus-TCP main websites, Modbus-TCP as the executable program under Linux, in executable file In the operation of Modbus-TCP slave station programs is completed by system () function；

3.2nd step, for POWERLINK main websites, POWERLINK slave stations program code is generated a Linux after cross compile Kernel module could run, it is necessary to be loaded into linux kernel by insmod orders；Here, by writing a shell pin Presents, by performing system () come perform script file in executable program, so as to complete POWERLINK slave station journeys The load operating of sequence.